1、SLIC（PAMI2012）
Title：SLIC Superpixels Compared to State-of-the-art Superpixel Methods
Author：Radhakrishna Achanta ... (École Polytechnique Fédérale de Lausanne,EPFL 瑞士聯邦理工學院)
Other Algorithms for generating superpixels
1.Graph-based algorithms
- treat each pixel as a node
- Edge weights are similarity between neighboring pixels.
- bipartite graph
- finding optimal paths
2.Gradient-ascent-based algorithms
算法:
Advantages
- Fastest
- most memory efficient
結果
1. 自然圖像
2. 2D and 3D EM images
2、Superpixel Sampling Networks(ECCV2018)
Title：Superpixel Sampling Networks
Author：Varun Jampani ... (NVIDIA)
Why is SLIC not differentiable?
- a non-differentiable nearest neighbor operation
- Associate each pixel to the nearest superpixel center
Advantages：
soft-associations
- the first end-to-end trainable superpixel algorithm
- convert the nearest-neighbor operation into differentiable
- learning with flexible loss functions
算法
- m：superpixel個數
- QF=weighted sum of pixel features，距離爲權值，對特徵加權
- Optional：求每個superpixel內的最大距離值，最小化這個值
- column normalized Qt as Qˆt
Loss function：
segmentation tasks： cross-entropy loss
optical flow ： L1-norm
compactness loss ：lower spatial variance
結果：
